Pretty doubtful about computer use/screenshotting based approaches.
With Retriever AI, we construct custom accessibility trees to represent web pages and just switched over to using DeepSeek v4 Flash and its nearing 100x cost decrease.
We also had great success just reverse engineering the underlying APIs of websites and then writing code to hit them. This approach of using screenshots to take actions on a webpage to trigger the underlying network calls the website is making seems too naive.
Fable has been gone almost a week. A god-tier coding model got banned with vague “national security” handwaving.
Trump administration friends get free passes on environmental laws, but anyone not pushing their narrative is getting curbstomped till they change their tune.
Come speak up against this regulatory retaliation and voice your support for Anthropic's Fable release.
We are meeting at a public plaza in front of their office to show support and discuss AI policy with free chai!
What do you think of just reverse engineering the network requests and writing scripts to hit the underlying APIs that the browser is making instead of tackling at the browser automation layer?
I am really wondering if Google is just subleasing Anthropic capacity. The terms are suspiciously the same as Anthropic's and Anthropic was supposed to have leased out all of Colossus 1.
Maybe with the corp token spending limits and the rise of codex, Anthropic saw steep deceleration in usuage?
The hard part about the web is that API's aren't just available even if the website owner wants them exposed (big if).
I embedded a Google Calendar widget on my Book a demo page, I don't know the API and Google doesn't expose/maintain one either.
What we are doing at Retriever AI is to instead reverse engineer the website APIs on the fly and call them directly from within the webpage so that auth/session tokens propoagate for free:
https://www.rtrvr.ai/blog/ai-subroutines-zero-token-determin...
there are no guarantees in life, can look up any random day and see a meteor streaking across the sky and realize that this is the end regardless of "sociopolitical structures".
All that matters is sociotechnological progress to be able to progress further enough to overcome these tests of existence.
Yes exactly imagine now anyone, even non-technical people, can just prompt and interact with this hidden/deeper layer of the web, all in their regular browser!
Hey thats a great idea, we will take a look into exploring this export option. But how would it save time by being a Playwright script?
Right now since we have a custom sandbox to re-execute the code in, we are using our own syntax and exposed methods. So even now you can edit the generated script.